Postoperative Glucose Control and SCIP Measures Gorav Ailawadi, MD Chief, Adult Cardiac Surgery University of Virginia April 25, 2015

Diabetes in CABG Incidence of Diabetes in cardiac surgery increased 85% over last 25 years 37% of Isolated CABG have Diabetes 50% of patients develop perioperative hyperglycemia

Hyperglycemia and Mortality Non-diabetics with hyperglycemia have higher mortality than those with controlled Diabetes Well known perioperative hyperglycemia increases mortality – Perioperative glycemic control improves mortality and morbidity

Outline 1.History of SCIP measures 2.Studies from our Center 3.Current Practice

SCIP Measures History Designed to decrease Infection/ Improve outcomes Developed metrics that hospitals must achieve : Post Cardiac Surgery POD 1 or 2 (0600) glucose < : Maintain glucose <180 in first day following surgery including : SCIP measures for glucose suspended

What is optimal glucose control?

Landmark Hyperglycemia Trial Intensive insulin therapy (Glucose  110 mg/dl) reduced mortality compared to Conventional (Glucose <215) in SICU Van den Berghe et al. N Engl J Med 2001

Intensive glucose control (Glucose mg/dl) had higher mortality versus Conventional (Glucose  180 mg/dl) NICE – SUGAR Trial Finfer et al. N Engl J Med 2009

Study 1 Is Tight Control Better than Moderate Control in Cardiac Surgery?

University of Virginia Isolated CABG ( ) N=8,662 Known Diabetes Mellitus (Type II) OR Developed perioperative hyperglycemia N=4,658 GROUP I Tight ≤126mg/dl N=134 GROUP II Moderate mg/dl N=2,785 GROUP III Liberal ≥180mg/dl N=1,739 First postoperative glucose >126 mg/dl Postoperative (3 day mean) glucose >126 mg/dl

Mortality *P<0.001

Major Complications *P=0.02

Hypoglycemia (1st Postop Glucose  60 mg/dl) * *P=0.02

Summary Moderate glucose control: – Less hypoglycemia – Less prolonged ventilation – 2 day shorter length of stay – 30% lower major complication rate – 40% lower odds of death

Study 2 Are SCIP measures really useful measure of outcomes?

Surgical Care Improvement Project (SCIP) Measure for Postoperative Glucose Control Should Not Be Used as a Measure of Quality Following Cardiac Surgery Damien J. LaPar MD, MSc, James M. Isbell MD, MSCI, John A. Kern MD, Gorav Ailawadi MD, Irving L. Kron MD Virginia Interdisciplinary Cardiothoracic Outcomes Research (VICTOR) Center 39 th Annual Meeting Western Thoracic Surgical Association Coeur d’Alene, ID Study 2

UVA: (n=1,703) Non-SCIP (BG>200) (n=176) SCIP (BG≤200) (n=1,527) 1:1 Propensity Matching (Nearest Neighbor Methodology) SCIP (BG≤200) (n=176) Non-SCIP (BG>200) (n=176)

POD 1 POD [207, 243] 135 [119, 154] 224 [210, 259] 131 [117, 150] Median Glucose Level

Variable SCIP (n=176) Non-SCIP (n=176) P Perioperative Myocardial Infarction0.00.6>0.99 Stroke Pneumonia Prolonged Ventilation Renal Failure Deep Sternal Wound Infection/Mediastinitis >0.99 Major Sternal Complication0.00.6>0.99 Composite Major Morbidity Operative Mortality Median Total ICU Length of Stay (Hours)52 [21,98]64 [23,116]0.61 Median Postoperative Length of Stay (Days) 6[4,9]6 [5,9] 0.08 Propensity Matched Cohort: NO DIFFERENCES

No Association SCIP Status ~ Adjusted Outcomes (CABG Only Patients) Outcome Odds Ratio or Regression Coefficient 95% C.I. or SE P Mortality , Major Morbidity , Major Sternal Complications--- Total ICU Duration (Hours) Postoperative Length of Stay (Days)

Achieving the SCIP measure for controlled POD 1 and 2 blood glucose is not associated with improved outcomes. Summary

Compliance with the new Surgical Care Improvement Project (SCIP) measure for glycemic control after cardiac surgery is not associated with improved postoperative outcomes James M. Isbell, MD, MSCI, Damien J. LaPar, MD, MSc, Robert H. Thiele, MD, John A. Kern, MD, Gorav Ailawadi, MD, Irving L. Kron, MD, Anthony L. McCall, MD, PhD, Jennifer L. Kirby, MD, PhD Western Thoracic Surgical Association Whistler, British Columbia June 24, 2015 To Be Presented at WTSA 2015!!!

Current Practice 2014: we instituted an inpatient CV diabetes consult service with a dedicated NP, aided by an attending endocrinologist Also initiated Glucommander- FDA approved device to aid in optimal BG control

Multidisciplinary Team Jay Isbell, MD Jennifer Kriby, MD Cherie Chaney, NP ICU Intensivist Endocrinologist CV Diabetes Consult

Privileged and Confidential Quality Assurance Document Privileged Under Virginia Code Section Cut Hyperglycemia by >50%

DWMBG = day-weight mean BG average BG for a patient over a day Mean Glucose Level

FLOOR

Conclusion SCIP measures currently suspended but likely to come back once validated Multidisciplinary approach yields optimal outcomes Midlevel provider on endocrinology/ diabetes service essential for education and excellent outcomes

Questions? Gorav Ailawadi